diff --git a/README.md b/README.md index d938bd8..405b214 100644 --- a/README.md +++ b/README.md @@ -38,7 +38,30 @@ more details, including on how to concatenate multi-part files. | Link | Type | Size/GB | Notes | |:-----|:-----|--------:|:------| +| [GGUF](https://huggingface.co/mradermacher/UltraLink-LM-i1-GGUF/resolve/main/UltraLink-LM.i1-IQ1_S.gguf) | i1-IQ1_S | 3.0 | for the desperate | +| [GGUF](https://huggingface.co/mradermacher/UltraLink-LM-i1-GGUF/resolve/main/UltraLink-LM.i1-IQ1_M.gguf) | i1-IQ1_M | 3.2 | mostly desperate | +| [GGUF](https://huggingface.co/mradermacher/UltraLink-LM-i1-GGUF/resolve/main/UltraLink-LM.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 3.6 | | +| [GGUF](https://huggingface.co/mradermacher/UltraLink-LM-i1-GGUF/resolve/main/UltraLink-LM.i1-IQ2_XS.gguf) | i1-IQ2_XS | 4.0 | | +| [GGUF](https://huggingface.co/mradermacher/UltraLink-LM-i1-GGUF/resolve/main/UltraLink-LM.i1-IQ2_S.gguf) | i1-IQ2_S | 4.3 | | +| [GGUF](https://huggingface.co/mradermacher/UltraLink-LM-i1-GGUF/resolve/main/UltraLink-LM.i1-IQ2_M.gguf) | i1-IQ2_M | 4.6 | | | [GGUF](https://huggingface.co/mradermacher/UltraLink-LM-i1-GGUF/resolve/main/UltraLink-LM.i1-Q2_K.gguf) | i1-Q2_K | 5.0 | IQ3_XXS probably better | +| [GGUF](https://huggingface.co/mradermacher/UltraLink-LM-i1-GGUF/resolve/main/UltraLink-LM.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 5.1 | lower quality | +| [GGUF](https://huggingface.co/mradermacher/UltraLink-LM-i1-GGUF/resolve/main/UltraLink-LM.i1-IQ3_XS.gguf) | i1-IQ3_XS | 5.5 | | +| [GGUF](https://huggingface.co/mradermacher/UltraLink-LM-i1-GGUF/resolve/main/UltraLink-LM.i1-IQ3_S.gguf) | i1-IQ3_S | 5.8 | beats Q3_K* | +| [GGUF](https://huggingface.co/mradermacher/UltraLink-LM-i1-GGUF/resolve/main/UltraLink-LM.i1-Q3_K_S.gguf) | i1-Q3_K_S | 5.8 | IQ3_XS probably better | +| [GGUF](https://huggingface.co/mradermacher/UltraLink-LM-i1-GGUF/resolve/main/UltraLink-LM.i1-IQ3_M.gguf) | i1-IQ3_M | 6.1 | | +| [GGUF](https://huggingface.co/mradermacher/UltraLink-LM-i1-GGUF/resolve/main/UltraLink-LM.i1-Q3_K_M.gguf) | i1-Q3_K_M | 6.4 | IQ3_S probably better | +| [GGUF](https://huggingface.co/mradermacher/UltraLink-LM-i1-GGUF/resolve/main/UltraLink-LM.i1-Q3_K_L.gguf) | i1-Q3_K_L | 7.0 | IQ3_M probably better | +| [GGUF](https://huggingface.co/mradermacher/UltraLink-LM-i1-GGUF/resolve/main/UltraLink-LM.i1-IQ4_XS.gguf) | i1-IQ4_XS | 7.1 | | +| [GGUF](https://huggingface.co/mradermacher/UltraLink-LM-i1-GGUF/resolve/main/UltraLink-LM.i1-Q4_0_4_4.gguf) | i1-Q4_0_4_4 | 7.5 | fast on arm, low quality | +| [GGUF](https://huggingface.co/mradermacher/UltraLink-LM-i1-GGUF/resolve/main/UltraLink-LM.i1-Q4_0_4_8.gguf) | i1-Q4_0_4_8 | 7.5 | fast on arm+i8mm, low quality | +| [GGUF](https://huggingface.co/mradermacher/UltraLink-LM-i1-GGUF/resolve/main/UltraLink-LM.i1-Q4_0_8_8.gguf) | i1-Q4_0_8_8 | 7.5 | fast on arm+sve, low quality | +| [GGUF](https://huggingface.co/mradermacher/UltraLink-LM-i1-GGUF/resolve/main/UltraLink-LM.i1-Q4_0.gguf) | i1-Q4_0 | 7.5 | fast, low quality | +| [GGUF](https://huggingface.co/mradermacher/UltraLink-LM-i1-GGUF/resolve/main/UltraLink-LM.i1-Q4_K_S.gguf) | i1-Q4_K_S | 7.5 | optimal size/speed/quality | +| [GGUF](https://huggingface.co/mradermacher/UltraLink-LM-i1-GGUF/resolve/main/UltraLink-LM.i1-Q4_K_M.gguf) | i1-Q4_K_M | 8.0 | fast, recommended | +| [GGUF](https://huggingface.co/mradermacher/UltraLink-LM-i1-GGUF/resolve/main/UltraLink-LM.i1-Q5_K_S.gguf) | i1-Q5_K_S | 9.1 | | +| [GGUF](https://huggingface.co/mradermacher/UltraLink-LM-i1-GGUF/resolve/main/UltraLink-LM.i1-Q5_K_M.gguf) | i1-Q5_K_M | 9.3 | | +| [GGUF](https://huggingface.co/mradermacher/UltraLink-LM-i1-GGUF/resolve/main/UltraLink-LM.i1-Q6_K.gguf) | i1-Q6_K | 10.8 | practically like static Q6_K | Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better):